Camera pose for moving vehicle

Abstract

A computer includes a processor and a memory, and the memory stores instructions executable by the processor to determine a first set of SLAM poses of a camera with respect to an environment by performing a simultaneous localization and mapping (SLAM) algorithm, determine a second set of G2O poses of the camera based on a plurality of ground-view images from the camera and an overhead image depicting the environment, and determine a third set of final poses of the camera by minimizing a loss function derived from a pose graph of the final poses. The loss function is based on the SLAM poses in the first set and the G2O poses in the second set.

What is claimed is: 
     
         1 . A computer comprising a processor and a memory, the memory storing instructions executable by the processor to:
 determine a first set of SLAM poses of a camera with respect to an environment by performing a simultaneous localization and mapping (SLAM) algorithm;   determine a second set of G 2 O poses of the camera based on a plurality of ground-view images from the camera and an overhead image depicting the environment; and   determine a third set of final poses of the camera by minimizing a loss function derived from a pose graph of the final poses, the loss function based on the SLAM poses in the first set and the G 2 O poses in the second set.   
     
     
         2 . The computer of  claim 1 , wherein the instructions further include instructions to actuate a component of a vehicle including the camera based on the final poses. 
     
     
         3 . The computer of  claim 1 , wherein the instructions further include instructions to, before determining the third set of the final poses, remove a first G 2 O pose from the second set upon determining that the first G 2 O pose is outside a spatial bound. 
     
     
         4 . The computer of  claim 3 , wherein the instructions further include instructions to determine the spatial bound based on the first set of the SLAM poses. 
     
     
         5 . The computer of  claim 3 , wherein the instructions further include instructions to determine the spatial bound based on an uncertainty measure of the first set of the SLAM poses. 
     
     
         6 . The computer of  claim 1 , wherein
 the first set includes a first SLAM pose at a first timestep and a second SLAM pose at a second timestep immediately following the first timestep;   the second set includes a first G 2 O pose at the first timestep and a second G 2 O pose at the second timestep; and   the instructions further include instructions to, before determining the third set of the final poses, remove the second G 2 O pose from the second set based on a comparison of a first change from the first SLAM pose to the second SLAM pose and a second change from the first G 2 O pose to the second G 2 O pose.   
     
     
         7 . The computer of  claim 6 , wherein the first change and the second change are rotations. 
     
     
         8 . The computer of  claim 6 , wherein the first change and the second change are translations. 
     
     
         9 . The computer of  claim 6 , wherein the instructions further include instructions to remove the second G 2 O pose from the second set in response to the comparison exceeding a threshold. 
     
     
         10 . The computer of  claim 1 , wherein the pose graph includes the final poses as graph nodes and a plurality of error terms as graph edges, and the loss function includes the error terms. 
     
     
         11 . The computer of  claim 10 , wherein the error terms include at least one term penalizing deviation between the final poses in the third set and the SLAM poses in the first set. 
     
     
         12 . The computer of  claim 11 , wherein the error terms include separate terms penalizing rotational deviation between the final poses in the third set and the SLAM poses in the first set and penalizing translational deviation between the final poses in the third set and the SLAM poses in the first set. 
     
     
         13 . The computer of  claim 10 , wherein the error terms include at least one term penalizing deviation between the final poses in the third set and the G 2 O poses in the second set. 
     
     
         14 . The computer of  claim 13 , wherein the error terms include separate terms penalizing rotational deviation between the final poses in the third set and the G 2 O poses in the second set and penalizing translational deviation between the final poses in the third set and the G 2 O poses in the second set. 
     
     
         15 . The computer of  claim 1 , wherein the SLAM poses, the G 2 O poses, and the final poses each include two spatial dimensions and one angular dimension. 
     
     
         16 . A method comprising:
 determining a first set of SLAM poses of a camera with respect to an environment by performing a simultaneous localization and mapping (SLAM) algorithm;   determining a second set of G 2 O poses of the camera based on a plurality of ground-view images from the camera and an overhead image depicting the environment; and   determining a third set of final poses of the camera by minimizing a loss function derived from a pose graph of the final poses, the loss function based on the SLAM poses in the first set and the G 2 O poses in the second set.   
     
     
         17 . The method of  claim 16 , further comprising actuating a component of a vehicle including the camera based on the final poses. 
     
     
         18 . The method of  claim 16 , further comprising, before determining the third set of the final poses, removing a first G 2 O pose from the second set upon determining that the first G 2 O pose is outside a spatial bound. 
     
     
         19 . The method of  claim 16 , wherein
 the first set includes a first SLAM pose at a first timestep and a second SLAM pose at a second timestep immediately following the first timestep; and   the second set includes a first G 2 O pose at the first timestep and a second G 2 O pose at the second timestep;   the method further comprising, before determining the third set of the final poses, removing the second G 2 O pose from the second set based on a comparison of a first change from the first SLAM pose to the second SLAM pose and a second change from the first G 2 O pose to the second G 2 O pose.   
     
     
         20 . The method of  claim 16 , wherein the pose graph includes the final poses as graph nodes and a plurality of error terms as graph edges, and the loss function includes the error terms.